Neural networks can be used to predict material properties of compounds and mixtures. Neural networks that were trained to predict clearing temperatures of liquid crystalline compounds. Both nematic and smectic liquid crystalline compounds were used separately to train artificial neural networks to predict clearing temperatures. Several thousand compounds were used in training of the networks. Independent sets of molecules will use to test the quality of the prediction. For nematic liquid crystals a standard deviation of 13 degrees was obtained in predicting the clearing temperatures of the test set.